Cryo-electron tomography (CET) is the only method available for nanoscale study of the overall architecture fo cells in their native (frozen-hydrated) state. It is the only method capable of identifying and determining the structure and orientation of macromolecules in their cellular context. Specimens for CET are initially prepared by vitreous freezing; they remain fully hydrated, and no chemical fixatives or stains are used. However, because only specimens thinner than a few hundred nanometers can be effectively imaged in the electron microscope, use of CET has largely been limited to isolated cell organelles or small cells such as bacteria. Eukaryotic cells have been imaged only in special cases, and only at their peripheries. In order to extend CET to a broad range of cells and tissue, we were the first to use cryo-ultramicrotomy (mechanical sectioning) to prepare tissue sections sufficiently thin for CET. However, we have found sectioning of frozen material to be a difficult and poorly reproducible technique, fraught with unavoidable artifacts. Therefore, we have introduced an entirely new approach: cryo-focused-ion-beam (cryo-FIB) milling of vitreously frozen specimens for CET. Other laboratories have since confirmed our ground-breaking proof-of-concept experiments. Our goal in this proposal is to develop cryo-FIB milling into a practical, reliable tool for biological CET, and to enable the researcher to capture regions of interest in the specimen by correlative light and electron microscopy.